ROAD SAFETY IN INDIA: CHALLENGES AND OPPORTUNITIES

8/13/2019 ROAD SAFETY IN INDIA: CHALLENGES AND OPPORTUNITIES

1/62


2/62

ROAD SAFETY IN INDIA: CHALLENGES AND OPPORTUNITIES

Dinesh MohanOmer TsimhoniMichael Sivak

Michael J. Flannagan

The University of MichiganTransportation Research Institute

Ann Arbor, Michigan 48109-2150

U.S.A.

Report No. UMTRI-2009-1January 2009


3/62

i

Technical Report Documentation Page1. Report No.

UMTRI-2009-12. Government Accession No. 3. Recipient s Catalog No.

5. Report Date

January 20094. Title and Subtitle

Road Safety in India: Challenges and Opportunities6. Performing Organization Code

3838187. Author(s)

Mohan, D., Tsimhoni, O., Sivak, M., and Flannagan, M.J.8. Performing Organization Report No.

UMTRI-2009-110. Work Unit no. (TRAIS)9. Performing Organization Name and Address

The University of MichiganTransportation Research Institute2901 Baxter RoadAnn Arbor, Michigan 48109-2150 U.S.A.

11. Contract or Grant No.

13. Type of Report and Period Covered12. Sponsoring Agency Name and Address

The University of MichiganStrategic Worldwide Transportation 2020 14. Sponsoring Agency Code15. Supplementary Notes

The current members of Strategic Worldwide Transportation 2020 includeArvinMeritor, Bendix, Bosch, Continental Automotive Systems, FIA Foundation forthe Automobile and Society, Ford Motor Company, General Motors, Nissan TechnicalCenter North America, and Toyota Motor Engineering and Manufacturing NorthAmerica. Information about Strategic Worldwide Transportation 2020 is available at:http://www.umich.edu/~umtriswt 16. Abstract

The present report was designed to analyze the traffic safety situation in India, andto identify countermeasures for areas in which the total harm caused by crashes can besubstantially and readily reduced. The report focuses on two aspects of traffic safety in

India: challenges and opportunities. The first part of the report provides acomprehensive analysis of the current traffic safety situation in India. It is pointed outin this analysis that fatality rates have increased both on highways and in urban areasduring the past few years. Theoretical models suggest that the number of fatalities inIndia is not likely to start to decline for many years to come unless new policies areimplemented. Based on the present analysis, the following six areas are identified ashaving potential for substantially reducing fatalities in India: (1) pedestrians and othernon-motorists in urban areas, (2) pedestrians, other non-motorists, and slow vehicles onhighways, (3) motorcycles and small cars in urban areas, (4) over-involvement of trucksand buses, (5) nighttime driving, and (6) wrong-way drivers on divided highways. Thesecond part of the report outlines several promising countermeasures for each of these

six areas. The third part of the report presents a brief comparison of major traffic safetychallenges in India and China.17. Key Words

India, road safety, transportation safety, traffic crashes,road fatalities, motorization, driving, countermeasures

18. Distribution Statement

Unlimited

19. Security Classification (of this report)

None20. Security Classification (of this page)

None21. No. of Pages

6122. Price


4/62

ii

ACKNOWLEDGMENTS

This research was supported by Strategic Worldwide Transportation 2020

(http://www.umich.edu/~umtriswt ). The current members of this research consortium are

ArvinMeritor, Bendix, Bosch, Continental Automotive Systems, FIA Foundation for the

Automobile and Society, Ford Motor Company, General Motors, Nissan Technical

Center North America, and Toyota Motor Engineering and Manufacturing North

America.

Dinesh Mohans contribution was prepared while he was a visiting research

scientist at UMTRI from the Transportation Research and Injury Prevention Programme,

Indian Institute of Technology Delhi, New Delhi, India.


5/62

iii

CONTENTS

ACKNOWLEDGMENTS...............................................................................................ii

CHALLENGES: ROAD SAFETY IN INDIA ................................................................1

Introduction...............................................................................................................1

Analysis of national data, trends, and distributions..................................................... 5

Age and sex distribution ......................................................................................9

Temporal factors................................................................................................11

Road users killed in traffic crashes..................................................................... 12

Analysis of state-level data ...................................................................................... 13

Intercity roads..........................................................................................................20

City data.................................................................................................................. 29Million-plus cities..............................................................................................29

Analysis for Delhi, Mumbai, and Kota...............................................................32

Conclusions............................................................................................................. 42

OPPORTUNITIES: PROMISING COUNTERMEASURES ........................................ 44

COMPARISON OF ROAD SAFETY IN INDIA AND CHINA.................................... 49

SUMMARY.................................................................................................................. 52

REFERENCES..............................................................................................................53


6/62

1

CHALLENGES: ROAD SAFETY IN INDIA

Introduction

According to official statistics, 105,725 people were killed in road traffic crashes in Indiain 2006 (NCRB, 2007). The situation in India has worsened in recent years. Traffic fatalities

increased by about 5% per year from 1980 to 2000, and since then have increased by about 8%

per year for the four years for which statistics are available (Figure 1). This is attributable partly

to an increase in the number of vehicles on the road, and partly to the absence of a coordinated

official policy to control the problem. The fatality rate has increased from 36 fatalities per

million persons in 1980 to 95 fatalities per million persons in 2006.

Figure 1. Traffic fatalities, 1970 through 2006.

The total motor vehicle population has increased from about 300,000 in 1951 to about

73,000,000 in 2004 (see Figure 2). (The actual number of motor vehicles in the country may be

20-30% lower, as registration procedures do not remove many of the out-of-service vehicles

from the records. This issue is dealt with in more detail in a later section.) The numbers and


7/62

2

percents of various types of vehicles registered (in 2004) and sold (in 2007) in India are given in

Table 1. Most notably, motorcycles are more than five times as numerous as cars. It is also

interesting that the total of buses, trucks, and other vehicles is similar in magnitude to the number

of cars. These proportions of vehicle types are different from those in the high-income countries

and can influence fatality rate patterns. In the U.S. in 2005, for example, passenger cars

constituted 66% of vehicles on the road; trucks, and vans constituted 30%; motorcycles were

only 3%; and buses 1%.

Figure 2. Registered vehicles, 1951 through 2004 (Department of Road Transport andHighways, 2008a).


8/62

3

Table 1Vehicle fleet by type.

Registered (2004)* Sold (2007)**Vehicle type Number

(million)Percent

Number

(million)Percent

Motorized two-wheeler 51.92 71 7.42 78

Car 9.45 13 1.27 13

Bus 0.77 1

Truck 3.75 50.48 5

Other 6.83 9 0.38 4Total 72.72 100 9.55 100

* Department of Road Transport and Highways (2008a).

** Society of Indian Automobile Manufacturers (2008).

The gross national income per capita in India in 2006 was US$ 820 (or US$ 2,460 in

purchasing power equivalent), while the corresponding values for the U.S. were US$ 44,710 and

44,070 (World Bank, 2008). The Indian economy has been growing at about 8% per year for the

past five years and is expected to grow at an average of 8-10% per year over the next five years

(Planning Commission, 2006). Car and motorized two-wheeler sales have averaged 16% and

9% annual growth rates, respectively, over the past five years. The trends are illustrated in

Figure 3. Although the recent global economic turmoil is likely to affect sales, we can expect

some level of sales growth to continue into the future. However, major increases in fuel prices

are likely to have a greater negative influence on sales of cars than on sales of motorized two-

wheelers.


9/62

4

Figure 3. Vehicle sales, 1972 through 2006 (Society of Indian Automobile Manufacturers,2008).

India has a rural road network of over 3,000,000 km, and urban roads total more than

250,000 km. The national highways, with a total length of 65,569 km, serve as the arterial

network across the country. Roads carry about 61% of the freight and 85% of the passenger

traffic. Highways total about 66,000 km (2% of all roads) and carry 40% of the road traffic. The

ongoing project of four-laning the 5,900-km Golden Quadrilateral connecting Delhi, Mumbai,

Chennai, and Kolkata is nearing completion. The ongoing four-laning of the 7,300-km North-

South East-West corridor is to be completed by December 2009. The National Highway

Development Programme, involving a total investment of US$ 55 billion up to 2012, has been

proposed for constructing 1,000 km of new expressways, six-laning 6,500 km of the four-lane

highways comprising the Golden Quadrilateral and certain other high-density stretches, four-

laning the Golden Quadrilateral and NS-EW corridors, four-laning 10,000 km of high-density

national highways, and upgrading 20,000 km of smaller rural roads into two-lane highways

(Committee on Infrastructure, 2008; Department of Road Transport and Highways, 2008b).


10/62

5

In urban areas, road development has been helped by a national urban renewal mission

program, through which the central government assists city governments with federal funds

(Ministry of Urban Development, 2006). With a combination of higher investments in urban and

rural road infrastructure, increasing sales of motor vehicles, and 7-10% growth rates in the

economy, the trend of increasing road traffic fatalities may be exacerbated if corrective policies

are not put in place on an urgent basis. The government of India has indicated its concern by

accepting the Report of the Committee on Road Safety and Traffic Management, which has

recommended that a National Road Safety Agency be established in India through specific

enabling legislation on road safety (Committee on Infrastructure, 2007). This proposal is under

consideration by the national government.

The goal of the present report is to identify major road safety problems in India and to

discuss countermeasures that have promise to address these specific road safety problems. A

similar report for the situation in China was recently completed (Zhang, Tsimhoni, Sivak, and

Flannagan, 2008). The first part of the present report examines the available crash data and

identifies a limited number of important areas that are especially characteristic of the current

road safety situation in India, and for which relatively specific countermeasures are available.

The second part discusses several promising countermeasures. In this discussion, we have

organized the treatment of countermeasures in terms of an analysis that describes the total harm

from road crashes as the product of three components: exposure, risk, and consequences (Thulin

and Nilsson, 1994; Sivak and Tsimhoni, 2008).

Analysis of national data, trends, and distributions

As indicated above, according to official statistics, 105,725 people were killed and452,922 people were injured in road traffic crashes in India in 2006 (NCRB, 2007). However, a

study done in Bangalore shows that while the number of traffic crash deaths recorded by the

police may be reasonably reliable, the total number of injuries is grossly underestimated

(Gururaj, 2006). According to that study, deaths were underestimated by 5% and the number


11/62


12/62

7

Figure 4. Traffic fatality rates, 1971 through 2006 (NCRB, 2007).

Figure 4 shows the trends for fatalities per million persons and per 10,000 vehicles in

India. Fatalities per million persons indicate the national health burden of traffic crashes. For

India, this ratio has been increasing continuously for the last four decades, reaching 98 in 2006.

The number of fatalities per 10,000 vehicles has decreased from about 80 in 1970 to 15 in 2004.

However, this reduction in fatalities per vehicle does not necessarily indicate improvement in

safety conditions on the road. As the per-capita availability of motor vehicles increases, fatalities

per vehicle always decrease (Smeed, 1972; Kopits and Cropper, 2005), and therefore, a decrease

in this ratio is not necessarily an indicator of road safety conditions. What matters is the risk per

trip, even at the individual level, as most trips are necessary for human activity. The need for

trips to work places, educational institutions, and destinations for socializing/shopping does not

change substantially with income levels. Although modes of transportation may change, the

number of trips per person remains approximately constant over time and income (Knoflacher,

2007). Therefore, traffic fatalities per unit population can be taken as a rough indicator of risk


13/62

8

faced by individuals. In this report, we use traffic fatalities per unit population as an indicator of

the health of the society at the city, region, or national level.

Figure 5 shows fatality rates in various countries as a function of per-capita income. The

Indian fatality rate (represented by the red square) is in the middle of the range for low-income

countries (Mohan, 2004). The fatality rates in mid-income countries range between half that of

India to five times greater. As incomes in India increase along with motor-vehicle use, the

present trends in India and the experience of some mid-income countries suggest that fatalities

could see a dramatic rise before they start to drop, consistent with the so-called Kuznets curve

(e.g., McManus, 2007). However, future trends may be altered if vehicle design, road building,

and traffic management policies include the latest scientific countermeasures.

Figure 5. Traffic fatality rates per 100,000 persons in 115 countries. India is represented by thered square.


14/62


15/62

10

Figure 7. Traffic fatalities by gender, 2006.

Figure 7 shows fatalities by age for males and females (NCRB, 2007). Among children,

the male fatality rate is greater than that of females by a factor of 2.8, whereas among older

groups the male-to-female ratio ranges between 4:1 and 6:1 (which may be partly explained by

the exposure of older groups on rural and urban roads). The primary and secondary school

enrollment ratios for males to females are reported to be 1.07:1 and 1.26:1, respectively

(UNICEF, 2007). However, this does not fully explain the difference. The male-to-female ratio

of adult literacy rate in India is 1.5:1 (UNICEF, 2007), and the male-to-female ratios of labor

force in rural and urban areas are 1.7:1 and 3.5:1, respectively (National Sample Survey

Organisation, 2005). This may partly account for the exposure of males and females on urban

and rural roads. In the U.S., by comparison, the fatal crash involvement rate per 100,000 persons

in 2006 was almost three times higher for male drivers than for females (NHTSA, 2008). Malesaccounted for 70% of all traffic fatalities, 69% of all pedestrian fatalities, and 88% of all bicyclist

fatalities in 2006. For all traffic fatalities, the male-to-female ratio of 2.3:1 in the U.S. is lower

than that in India. Higher female participation in the work force and presence on the roads may

partly account for this difference.


16/62

11

Temporal factors

Figure 8 shows fatalities by month and Figure 9 shows fatalities by time of day (NCRB,

2007). For the country as a whole, monthly variations are not substantial, but hourly variations

are. The fatalities remain relatively constant and high during working hours, and low during

early hours of the morning. However, no detailed information is available to analyze the

contributing factors.

Figure 8. Traffic fatalities by month, 2006.


17/62

12

Figure 9. Traffic fatalities by time of day, 2006.

Road users killed in traffic crashes

Data for road user types killed are not available at the national or state levels in India.

Some cities maintain such details locally, but data are not available for all cities in the country.

Proportions of road users killed in the late 1990s in the cities of Mumbai and Delhi and selected

highway locations are given in Table 2 (Mohan and Tiwari, 2000). These data show that car

occupants were a small proportion of the total fatalities. Pedestrians, bicyclists, and motorized

two-wheeler riders accounted for 60-90% of all traffic fatalities. This pattern is very different

from that obtained in all high-income countries.


18/62

13

Table 2Traffic fatalities by road user and type.

Location (percent)Type of road user

Mumbai Delhi Highways*

Truck 2 2 14Bus 1 5 3

Car 2 3 15

Three-wheeled scooter taxi 4 3 -

Motorized two-wheeler 7 21 24

Human and animal powered vehicle 0 3 1

Bicycle 6 10 11

Pedestrian 78 53 32

Total 100 100 100* The data are for 14 selected locations, and thus might not be representative for the

entire country. (Tractor fatalities are not included.)

Analysis of state-level data

India is divided into 28 states and 7 union territories, the latter being administered by the

central government. Among the states, Sikkim has the smallest population (540,493) and Goa

the smallest area (3,702 km 2), whereas Uttar Pradesh has the largest population (166,052,859)

and Rajasthan the largest area (342,239 km 2). The union territories are relatively smaller:

Lakshadweep has the smallest population (60,595) and area (32 km 2), Delhi (capital of India) has

the largest population (13,782,976), and Andaman and Nicobar Islands the largest area (8,249

km 2). Among the states, Arunachal Pradesh (13 persons per km 2) and West Bengal (904 persons

per km 2), and among the union territories, Andaman and Nicobar Islands (43 persons per km 2)

and Delhi (9,294 persons per km 2) have the lowest and highest population densities (Registrar

General & Consensus Commissioner India, 2008; all statistics are for 2001). These data show

that the various states have widely varying characteristics.


19/62


20/62

15

Figure 10 shows the total number of fatalities by state and territory from 1991 to 2006.

Manipur, Meghalaya, Mizoram, Nagaland, and Tripura are small states with hilly terrain.

Andaman and Nicobar Islands, Dadra and Nagar Haveli, Daman and Diu, Lakshadweep,

Puducherry, Chandigarh and Delhi are union territories, which are generally small, and the last

two are cities. Therefore, these regions can have different traffic and fatality patterns.

Regardless, the total number of fatalities increased by more than 70% in most regions between

1991 and 2006.

Figure 11 shows the fatalities per million population and Figure 12 the ratios between

fatalities per million population in 2006 and 1991. Fatality rates per million population also

increased in most regions except in the northeastern hilly states and the city of Delhi. The

increase was 50% or more in 11 states and union territories. Figure 13 shows the fatality rate

(per million persons) as a function of the population density in states and territories. There does

not seem to be a strong correlation of fatality rates with population density.

Figure 14 shows the association between fatalities per 100,000 persons (2003) and

income per-capita of states (expressed in 2003 US$) (Ministry of Statistics and Programme

Implementation, 2008). Figure 15 shows the relationship between increasing incomes and

vehicle ownership. These data show that states with higher per-capita incomes generally have

higher vehicle ownership and higher fatality rates. Many policy makers identify this increasing

vehicle ownership as the cause for higher fatality rates. However, increasing vehicle ownership

need not result in increased fatality rates if adequate safety measures are implemented.


21/62


22/62

17

Figure 12. Change in traffic fatalities per population from 1991 to 2006 by state and territory.


23/62


24/62

19

Figure 15. Vehicle ownership vs. income per capita for states and territories, 2003 (Departmentof Road Transport and Highways, 2008a; Ministry of Statistics and Programme Implementation,2008).

Many countries with higher vehicle ownership rates than India have lower fatality rates

(Table 3), but all of these countries with lower fatality rates have high incomes (IRF, 2007).

Kopits and Cropper (2005) suggest that the income per-capita at which the motor vehicle fatality

rate begins to decline is approximately $6,100 (1985 international dollars) when a common time

trend is assumed for all countries, and $8,600 when separate time trends are used for each

geographic region. Koornstra (2007), using a cyclically modulated risk decay function model,

predicts that countries like India are unlikely to see declining rates before 2030. The income per-

capita in India was estimated at US$ 820 in 2006 (World Bank, 2008). At an average growth

rate of 8% per year, the per capita income of India in 2030 would be US$ 5,200. Therefore, ifwe depend on current traffic management and road design policies, Indian states can expect

increasing death rates into the foreseeable future according to both scenarios. This is clearly not

desirable.


25/62

20

Table 3Fatality rates in selected countries, 2005.

CountryVehicles per 100

personsFatalities per

100,000 persons

Sweden 51 4.9 Netherlands 48 4.9

United Kingdom 51 5.3

Switzerland 56 5.5

Norway 55 5.6

Denmark 44 6.1

Iceland 72 6.4

Germany 58 6.5

Japan 59 6.6

India* 7 9.5

* Estimates for 2006

Safety policies depend on changes in vehicle design, road infrastructure, and enforcement

to bring down injury and fatality rates. Because car occupants are a small proportion of all

fatalities, and are likely to remain so in the near future, improvements in car design are not likely

to have as large an effect on overall fatalities as they had in high-income countries in the past 30

years. In any event, a large proportion of cars sold in India already conform to international

standards, and safety regulations are expected to be fully aligned with the ECE by 2010

(Government of India, 2006).

Intercity roads

India has a total road network of 3.3 million km. The distribution of various categories of

roads is shown in Table 4, and the share of national highways by road length is shown in Table 5

(Department of Road Transport and Highways, 2008c). The total length of access-controlled

highways is less than 600 km at present. Although national highways constitute only about 2%

of the total road network, about 40% of the total road traffic moves on these roads. The Ministry

of Road Transport and Highways administers the national highway system, while state highways


26/62

21

and other state roads are maintained by state public works departments, and other, more local

roads are maintained by municipalities, districts, and villages.

Table 4Total length of roads by category.

Road categoryTotal length

(km)

National highway 66,754

State highway 128,000

District road 470,000

Rural road 2,650,000

Total 3,314,754

Table 5Total length of national highways by width.

Highway width Total length (km) Percent

1 lane, intermediate lane 18,350 27

2 lanes 39,079 59

4-8 lanes 9,325 14

Figure 16 shows the proportions of fatalities on different categories of roads in Indian

states. This figure shows that about two-thirds of the fatalities take place on national and state

highways, which are mainly intercity roads (Ministry of Shipping, Road Transport and

Highways, 2006). Fatality rates per 100 km of national highway in 2004 for states with largely

non-hilly terrain are plotted in Figure 17 as a function of income per capita (Ministry of

Shipping, Road Transport and Highways, 2006; Department of Road Transport and Highways,

2008c; Ministry of Statistics and Programme Implementation, 2008). Table 6 shows the fatality

rates for states with hilly terrain. The average fatality rate for national highways in non-hilly

states was 59 persons per km per year in 2004, whereas the same rate in hilly states was 16.


27/62

22

Figure 16. Percentages of traffic fatalities by road category and state or territory, 2004.


28/62

23

Figure 17. Traffic fatality rates on national highways in states with non-hilly terrain vs. income per capita, 2004.

Table 6Fatalities on national highways in hilly states, 2004.

StatesFatalities,number Length (km)

Fatalities per 100 km

Jammu & Kashmir 350 823 43

Himachal Pradesh 280 1,208 23

Uttaranchal 425 1,991 21

Arunachal Pradesh 67 392 17

Meghalaya 87 810 11

Nagaland 36 494 7

Manipur 53 959 6Mizoram 36 927 4

Mean 16


29/62

24

Detailed data are not available at the national or state level for crashes on national

highways. A study sponsored by the Ministry of Road Transport and Highways collected data

on modal shares, vehicle speeds, and traffic crashes on selected locations on national and state

highways around the country in the late 1990s (Tiwari, Mohan, and Gupta, 2000). Figure 18

shows the type of road users killed on highways and the impacting vehicles, respectively. Two

findings are of particular interest. First, the fatalities in this sample included 32% of motor-

vehicle occupants and 68% of vulnerable road users. Though the motor vehicle fatalities were

higher on highways than in urban areas, as would be expected, their proportion was not as high

as in high-income countries. Second, trucks were the striking party in 65% of fatal crashes.

Figure 18. Traffic fatalities and striking vehicle by road-user type on national highways, 1999.

Table 7 shows the crash statistics by categories of highways (Tiwari et al., 2000). There

were no major differences in rollover crashes on two-lane and four-lane roads. Similarly, there

were no major differences in head-on collisions on different types of two-lane roads. However,


30/62


31/62

26

Venkatramayya, and Kashinath have done a study of crashes on the Dindigul-Palani section of

National Highway 209 and report that about 50% of the crashes involved buses, 25% of the

victims were pedestrians, and two stretches of the highway had a higher number of crashes than

other sections (Kumar and Venkatramayya, 2004). A study of crashes on National Highway 8

passing through Valsad District found that crashes were increasing at a rate of 3.9%, rear-end

crashes were 40% of total crashes, and heavy vehicles were involved in the largest number of

cases (Saija and Patel, 2002). These studies inform us that highways do have some stretches that

can be identified as being associated with a higher number of crashes than other locations, heavy

vehicles are involved in a larger number of crashes than lighter vehicles, and vulnerable road

users are a significant proportion of those killed on national highways. However, none of these

studies provides information on speeds, modal shares, and highway design or their association

with road traffic fatalities.

Shaheem et al. have published two detailed studies on road traffic crashes on the Aluva-

Cherthala and Pallichal-Kaliyikkavila sections of National Highway 47 in Kerala (Shaheem and

Das Gupta, 2005; Shaheem, Mohammed, and Rajeevan, 2006). For the Pallichal-Kaliyikkavila

section, the authors evaluated the impact of four-laning of 38.5 km of the highway on road traffic

crashes. They also reported that heavy vehicles had a high involvement, and pedestrians and

cyclists were 28% of the victims. The most important finding of this study is that the fatality rate

per volume is more than three times higher on the four-lane section than on two-lane sections.

The fatality rate based on population density of the associated regions was higher on the four-

lane section compared to two-lane sections, and conversion of two-lane sections to four-lane

sections increased the fatality rate from 41% to 51% on the high-crash-rate sections.

A recent study of crashes in the state of Punjab reported that 30% of the crashes took

place in urban areas, with the rest in semi-urban and rural areas (Consulting Engineering

Services (I) Pvt. Ltd, 2007). The crash rate was 2.4 fatalities/km on national highways and 0.9

on state highways. The distribution of fatal crashes on rural roads is given in Table 8. These

figures include all kinds of roads in rural areas and so are not directly comparable with the data


32/62

27

in Table 7. Bicycle crashes have not been reported separately and the definitions used for

various categories would be different. However, two issues are similar: the incidence of rear-end

crashes and the involvement of pedestrians on highways.

Table 8Fatal crashes by type on rural roads in Punjab.

Crash type Percent

Rollover 1.0

Head-on 31.2

Rear-end 11.8

Side swipe 1.1

Right angle 15.2Pedestrian 13.0

Others 26.7Total 100.0

Various studies suggest high involvement of trucks in fatal crashes on rural highways in

India, but only one study quantified this (Tiwari et al., 2000). Figure 19 shows average modal

shares of road users in 2006 obtained from the data collected by the Ministry of Road Transport

and Highways at sample stations on national highways in different states (Department of Road

Transport and Highways, 2008d). The average values for all states were as follows: 33% for cars

and three-wheeled vehicles, 29% for motorized two-wheelers, 29% for trucks, 7% for buses, and

2% for tractors. These modal shares are different from those in high-income countries. This

partly explains the high involvement of trucks and motorized two wheelers in crashes on Indian

rural highways. Accurate estimates for non-motorized road user share on Indian roads are notavailable, but the data from the study done in late 1990s gives the following ranges: 4-25% on

four-lane divided highways and 9-39% on two-lane highways (Tiwari et al., 2000). The

presence of pedestrians and bicyclists in significant numbers on highways accounts for the high

fatality rates for these groups, even on intercity roads.


33/62

28

Figure 19. Percentages of road users on national highways by state and territory, 2006.

In summary, it is clear that crash rates on intercity roads are high. The construction of

four-lane divided highways (without access control) does not seem to have reduced fatality rates,

and vulnerable road users still account for a large proportion of fatalities. The presence of slow

modes on highways creates serious problems, as speed differentials can account for significant

increases in crash rates (Koornstra, 2007). High incidence of fatal rear-end crashes suggests a

general lack of visibility with a possible contribution of poor conspicuity of parked vehicles.

There is a clear case for redesign of intercity roads with separation of slow and fast modes. The

needs of road users on local short distance trips will have to be accounted for to reduce the

probability of head-on crashes (due to drivers going the wrong way on divided highways) by

provision of continuous service lanes and safe road crossings at convenient distances. Solutions


34/62

29

for many of these issues are not readily available, and research studies are necessary for the

evolution of new designs.

City data

Million-plus cities

According to the 2001 census, 35 cities in India have populations of 1 million or more.

Data for the total number of fatalities are available for these cities and are shown in Figure 20

(NCRB, 2007). Delhi had the highest number of fatalities in 2006 (1,794) with a rate of 140 per

million population. The lowest rate was in Amritsar (33) and the highest in Agra (317), with an

overall average of 142 fatalities per million persons for all these cities. In comparison, there

were 245 cities with populations greater than 100,000 persons in the U.S. in 1998. For these

cities, the average fatality rate was 98 per million persons, with Dallas having the highest rate

(142) and New York the lowest (46). The average rate for million-plus cities is 45% higher in

India than in the U.S., with the highest rate being 2.2 times higher.

A comparison of the fatality rates per million persons between 2001 and 2006 is shown in

Figure 21. In this period of five years, only eight of the 35 cities did not experience an increasein fatality rates. The highest increase was 550% in Asansol, with an overall average for all cities

of 5.5% compound per year. This is quite an alarming situation, as at this rate the city death rate

per million persons will be doubled in 12 years. Since a vast majority of the victims in these

cities are vulnerable road users, one possible cause could be increases in vehicle speeds. The

probability of pedestrian death is estimated at less than 10% at impact speeds of 30 km/h and

greater than 80% at 50 km/h, and the relationship between increase in fatalities and increase in

impact velocities is governed by a power of four (Leaf and Preusser, 1999; Koornstra, 2007).

Small increases in urban speeds can increase death rates dramatically.


35/62

30

Figure 20. Traffic fatalities in cities with populations of at least one million, 2006.


36/62


37/62


38/62


39/62

34

Finally, a complete dataset of fatalities from the city of Kota, Rajasthan were obtained

from the Kota police department for 2007. Kota, with a population of about 780,000, is

substantially smaller than Delhi or Mumbai. Although fewer fatalities occur in Kota, an analysis

of the situation in Kota adds a different perspective that may be useful for making inferences

about smaller cities with lower car ownership and much higher motorcycle ownership than

megacities.

Figure 23 shows the distribution of fatalities by road user in the three cities. It excludes

fatalities for which no information was available and focuses on road user fatalities that were

influenced by another road user. For example, it includes pedestrians that were hit by various

types of other road users, and includes crashes involving non-motorized and motorized vehicles

that were hit by another road user and that resulted in a fatality.

In all three cities, a majority of the fatalities were pedestrians, bicyclists, other non-motor

road users, and riders of motorized two-wheelers: 86% in Delhi, 93% in Mumbai, and 67% in

Kota. Mumbai differs substantially in having fewer fatalities among riders of motorized two-

wheelers. This may be partly because Mumbai has a much lower ownership rate of motorized

two-wheelers. Kota appears to have a much higher proportion of car involvement, but this is

misleading because for Kota the category cars included all kinds of shared-route taxis that

consist of cars, jeeps, and three- and four-wheeled locally manufactured vehicles. These taxis

carry a much larger number of passengers than can be accommodated by the available seats.

This practice ensures low, affordable fares and is common in smaller cities of India, as organized

public transport is not available. It should be noted that private cars accounted for only 5.5% of

the total number of fatalities, and 17% of those listed in the car group. Thus, occupants of

private cars constitute a very small proportion of fatalities, and people outside the car constitute

the vast majority of fatalities in all cities irrespective of size. Only the relative proportions of

pedestrians, bicyclists, and motorized two-wheeler fatalities differ, depending on city

characteristics.


40/62

35

Figure 23. Fatalities by road user in Delhi (2001-2005), Mumbai (1996-1997), and Kota (2007).


41/62

36

Figure 24 examines the types of road users that were the striking components of the

crashes in each of the three cities. Trucks and buses together are associated with more than 40%

of the fatalities in urban areas in all three cities. This is in comparison with 14% in the U.S.

(FARS, 2008). Buses are a smaller component in Kota, which has very limited public bus

service. Buses are replaced by jeep taxis (included in the car category) and other vehicles

transporting passengers as route taxis. The other difference between Kota and the large cities is

the involvement of tractors in crashes. This is probably because small cities have greater

business involvement with the surrounding rural areas, and farmers bring their produce for sale

into the city, using tractors as personal transport. The high involvement of trucks is due to the

fact that intercity highways go through populated parts of the city and are not separated from the

rest of the road users. Large cities can restrict passage of trucks through the city in the daytime,

but this is not possible in the smaller cities. The high involvement of buses and trucks has been

investigated in a detailed conflict-analysis study (Tiwari, Mohan, and Fazio, 1998). The authors

report that these vehicles have to use the curbside lane in the city and often come into conflict

with pedestrians, bicyclists, and motorized two-wheelers, as they are present in the same space

on the road. This results in frequent involvement of vulnerable road users in fatal crashes

because of the absence of dedicated bicycle paths and adequate space for pedestrian movement.

The involvement of motorized two-wheelers, though not very high, is significant because

it can be associated with pedestrian, bicyclist, and other motorcycle fatalities. Of particular

interest is the pattern of fatalities in which the striking vehicle is classified as unknown. Many

such cases are hit-and-run incidents. This is probably because the presence of traffic police is

limited at night and offenders can get away undetected. This is particularly true for Delhi. Since

it is more difficult for buses and trucks to escape after a crash than it is for cars, it is possible that

many in the unknown category could be the latter. If that is true, then the involvement of cars

would be higher than is shown in the figure. This also suggests that policing of roads should be

done around the clock, for more complete data to be available and for control of drivers.


42/62

37

Figure 24. Striking vehicle in fatal crashes in Delhi (2001-2005), Mumbai (1996-1997), andKota (2007).

Figure 25 shows involvement of different categories of road users by time of day in

Delhi. There are four main features of the data. First, despite the fact that nighttime exposure is

likely to be substantially lower than daytime exposure, nighttime crashes account for a large

proportion of fatalities. Second, trucks have high involvement in both daytime and nighttime.

Third, buses feature prominently from about 07:00 until about 21:00. Fourth, the proportion of

unknowns is substantial, especially during nighttime.


43/62

38

Figure 25. Fatalities by striking vehicle and time of day in Delhi, 2001-2005.

Figures 26 and 27 present the distribution of fatalities in Delhi as a function of time of

day and month of the year. The seasonal pattern of pedestrians, bicyclists, and non-motorized

vehicles is somewhat different from that of motorized vehicles. While the peaks between 10:00

and 15:00 are similar for motorized and non-motorized users, the nighttime peaks are not.

Motorized users have a relatively high number of crashes between 22:00 and 01:00, while non-

motorized users do not. This could be because of the differential exposure of these two groups at

night, higher speeds of vehicles at night due to low density, and/or higher frequency of driving

under the influence of alcohol at night. Evidence for increased use of alcohol comes from a

hospital study in Delhi where 29% of the riders of motorized two-wheelers admitted to alcohol

consumption before the crash (Mishra, Banerji, and Mohan, 1984). In Bangalore, a hospital-

based study showed that alcohol was involved in 22% of nighttime crashes, and that 35% ofrandomly checked drivers on the road at night were under the influence of alcohol (Gururaj,

2006).


44/62

39

Figure 26. Fatalities in motorized vehicles by month and time of day in Delhi, 2001-2005.

Figure 27. Fatalities of unprotected road users by month and time of day in Delhi, 2001-2005.


45/62

40

Another difference between the motorized and non-motorized users is the increase in

fatalities among the non-motorized users during the winter months between 17:00 and 19:00,

which is not present for the motorized road users. In Delhi, sunrise and sunset are at around

05:00 and 18:30, respectively, in midsummer, and at 07:00 and 17:30 in mid-winter. These

higher fatality rates could be due to the lack of conspicuity of pedestrians and bicyclists when the

sun starts setting early in the winter.

Figure 28 shows the distribution of all traffic fatalities in Delhi and Kota and pedestrian

fatalities in Mumbai, and their representation in the population. As shown earlier, in the data for

all of India, children are underrepresented in all three cities, whereas pedestrians in Mumbai are

overrepresented for ages greater than 45 years. The data are shown with reference to location in

Figure 29. The fact that a vast majority of crashes take place mid block and involve adults of all

ages suggests that vulnerable road users have a high possibility of conflict with motor vehicles

and that the velocities of these vehicles are high enough to cause fatalities. Therefore, urban

safety policies should give the highest priority to pedestrian and bicycle separation from motor

vehicles, speed control on main arterial roads, and traffic calming on all other roads.


46/62

41

Figure 28. Traffic fatalities by age group in Kota (2007; all traffic fatalities) and Mumbai (1996-1997; pedestrians only).

Figure 29. Locations of fatal crashes in Kota (2007) and Mumbai (1996-1997).


47/62


48/62

43

(1) Pedestrians, bicyclists, and other non-motorists in urban areas. This group of road

users currently accounts for about 60% of all fatalities in urban areas, substantially more than in

most high-income countries.

(2) Pedestrians, other non-motorists, and slow vehicles on national highways . Evidence

suggests that a high percentage (about 20-40%) of fatalities on highways consist of pedestrians,

bicyclists, other non-motorists, and occupants of slow vehicles. The problems associated with

pedestrians and other non-motorists on highways are inherently different from the problems in

urban areas. Therefore, it is appropriate to deal with these two problems separately.

(3) Motorcycles and small cars in urban areas. Motorcyclists represent a large portion

of urban fatalities (about 25%). The expected partial shift of motorcyclists to small cars is of

concern, because although small cars provide more protection to the occupants, they are

expected to be more harmful than motorcycles to pedestrians, bicyclists, and other motorcyclists

unless vehicle fronts are designed to be more forgiving.

(4) Over-involvement of trucks and buses in fatal crashes. Several studies indicate that

the involvement of trucks in fatal crashes is greater than would be expected based only on their

exposure.

(5) Nighttime driving: visibility, alcohol, and fatigue. Evidence suggests that, as is the

case in other countries, nighttime driving in India is substantially riskier than daytime driving.

Three aspects are of relevance here: conspicuity of road users, driving under the influence of

alcohol, and fatigue of truck drivers. The available data do not allow us to quantify the

individual contribution of each of these aspects.

(6) Wrong-way drivers on divided highways. A large proportion of fatalities on divided

highways are from head-on collisions. One study found that such crashes accounted for 19% of

all fatalities on four-lane, divided highways.


49/62

44

OPPORTUNITIES: PROMISING COUNTERMEASURES

In the first part of this report, we identified six areas that, with appropriate

countermeasures, are likely to bring about substantial improvements in road safety in India.

These areas are (1) pedestrians and other non-motorists in urban areas, (2) pedestrians, other

non-motorists, and slow vehicles on national highways, (3) motorcyclists and small cars in urban

areas, (4) over-involvement of trucks and buses in crashes, (5) nighttime driving, and (6) wrong-

way drivers on divided highways. In this part of the report, we will outline several safety

countermeasures for each of these areas.

General approach to identification of potential countermeasures

The approach that we used in this report is based on the concept of total harm (Thulin and

Nilsson, 1994; Sivak and Tsimhoni, 2008, Zhang et al., 2008). As we discussed in Sivak and

Tsimhoni (2008, p. 453),

total harm is conceptualized as a product of exposure, risk, and consequences.Exposure is the probability of a particular, potentially risky recent event(condition, situation) per distance traveled or per unit of time. Risk is theconditional probability of a crash, given the event in question. Consequence isthe conditional probability of undesirable outcomes (i.e., fatality, injury, and

property damage) given a crash that was precipitated by the event in question.For each event, the values along the three dimensions (exposure, risk, andconsequences) define a three-dimensional space. The volume of this space is thetotal harm for this particular event.

The goal of traffic-safety countermeasures should be a reduction in total harm, based on

influence on exposure, risk, or consequences.

Proposed countermeasures for India

Table 9 lists broad areas of intervention for each of the six problem areas discussed

above. The countermeasures are classified by three components that contribute to total harm

(exposure, risk, and consequences). Some countermeasures are known and have international


50/62


51/62

46

Table 9Summary of promising countermeasures.

Problem area Exposure Risk Consequences

Pedestrians andother non-motoristsin urban areas

Separation ofmotorized and non-motorized traffic onarterial roads

Speed controlRoundabouts

Restrictions onfree left turns

Pedestrian-friendlyfront ends of vehicles

Pedestrians, othernon-motorists, andslow vehicles onhighways

Special facilities forslow and localtraffic all alonghighways

Pedestriandetectiontechnology

Forward collisionwarning systems

Pedestrian-friendlyfront ends of vehicles

Crashworthiness ofslow vehicles

Motorcycles andsmall cars in urbanareas

Daytime runninglights

Improved lightingand signaling

Enforcement ofhelmet-use andseatbelt laws

Introduction of passivemeasures like airbags

Pedestrian/motorcycleimpact standards forsmall cars

Over-involvementof trucks and buses

Speed control

Rest regulations

for truck driversImproved vehicleconspicuity

Safer vehicle fronts

Nighttime driving

Improvedheadlighting

Improved vehicleconspicuity

Random alcoholbreath testing

Rest regulationsfor truck drivers

Wrong-way driverson dividedhighways

Highway designsincorporating localneeds

Enforcement

Forward collisionwarning systems


52/62

47

Pedestrians and other non-motorists in urban areas . The countermeasures with high

potential for major positive effects include separation of traffic on arterial roads and traffic

calming in all other areas (starting in urban areas that have the highest exposure of non-motorists

to other traffic), and improvement in the pedestrian friendliness of vehicle front ends.

International guidelines for traffic calming will have to be modified to incorporate designs that

are effective for motorcycles as well. Speed control, use of scientifically designed roundabouts

instead of traffic lights, and restrictions on free left turns (India drives on the left) are

internationally used measures that reduce urban road traffic injuries. Because of the expected

future partial shift in India from motorized two-wheelers to small cars (Cather, 2007), making

small cars more pedestrian friendly will be of increased importance

Pedestrians, other non-motorists, and slow vehicles on highways . The etiology of these

crashes is somewhat different from the analogous crashes in urban areas. These crashes involve

conditions with higher speeds and relatively lower frequency of exposure to non-motorists.

Highway designs in India will have to be modified to separate slow vehicles and pedestrians all

along the highway, and provide convenient road crossing facilities at frequent intervals for local

traffic. The recommended focus for the future is on in-vehicle pedestrian-detection technology

and pedestrian-friendly front ends of vehicles (for crashes involving non-motorists), and forward

collision warning systems and improved crashworthiness of slow vehicles (for crashes of slow

vehicles with other vehicles).

Motorcycles and small cars in urban areas . The following countermeasures are

recommended: required daytime running lights for motorcycles, improved lighting and signaling

on motorcycles, enforcement of motorcyclist helmet-use laws, and motorcycle-friendly front

ends of vehicles (for crashes involving motorcycles), and improved crashworthiness of small

cars (for crashes of small cars). Passive measures, like mandatory airbags for all cars, may prove

to be cost effective when enforcement measures are lacking.

Over-involvement of trucks and buses . Two countermeasures, aimed at the driving

performance of truck drivers, are recommended: speed control by use of data loggers and GPS


53/62

48

systems, and implementation of rest regulations. Also recommended, for both trucks and buses,

are safer vehicle fronts and improved vehicle conspicuity.

Nighttime driving . The available data do not allow quantification of the relative

contribution of visibility, alcohol, and fatigue to the increased risk of nighttime driving in India.

Consequently, the recommendation is to address all of these mechanisms with a combination of

countermeasures: improved headlighting and conspicuity of all vehicles (including bicycles and

other non-motorized vehicles), frequent and sustained random breath testing programs, and rest

regulations for truck drivers.

Wrong-way drivers on divided highways . Again, the available data are not sufficiently

detailed to provide information about the mechanisms that contribute to this type of crash.

Nevertheless, anecdotal evidence suggests that these crashes often involve drivers who take

shortcuts, often with slow vehicles such as farming equipment. In the short term, enforcement is

the key approach for addressing this phenomenon. Research needs to be undertaken to

understand the needs of local traffic and to develop standards for safer road crossings and the

frequency at which they need to be provided. Additionally, collision warning systems would

also contribute to a reduction of this type of crash.

As indicated above, detailed data are not available to suggest specific measures for all

road safety issues in the country. Therefore, there is an urgent need to revamp police data

collecting procedures so that necessary information is available for scientific analysis. Initially,

this effort may be focused on all fatal crashes. Statistically valid conclusions are also difficult

because a significant number of the crashes are classified as hit and run. Information on such

cases will be available only if enforcement is maintained twenty-four hours a day.


54/62

49

COMPARISON OF ROAD SAFETY IN INDIA AND CHINA

In a recent report on road safety in China (Zhang, Tsimhoni, Sivak, and Flannagan,

2008), we applied a similar approach for identification of potential countermeasures to reduce

the number of traffic fatalities in China. India and China have the two largest populations in the

world. The two countries have comparable social characteristics but different political

structures. Table 10 provides a comparison of several relevant aspects related to traffic safety in

India and China (Cather, 2007; CRTAS, 2007; Stevenson et al., 2007; Zhang et al., 2008). The

most notable differences are that India has lower income per capita, proportionally more

motorized two-wheelers, a higher fatality rate per population, more older people among thefatalities, proportionally fewer females among the fatalities, and a less advanced network of

access-controlled expressways. Table 11 summarizes the major traffic safety problems in India

(as discussed in this report) and in China (Zhang et al., 2008).

The information in Tables 10 and 11 indicates that the road-safety situations in India and

China are similar. This is the case in spite of the fact that income per capita in China is roughly

two and a half times greater than in India. Because it takes about 10 years for income to double

at a sustained growth rate of 7% per year, this comparison highlights the fact that road traffic

fatalities cannot be expected to decrease rapidly unless focused efforts with a sound scientific

base are put in place with a sense of urgency.


55/62

50

Table 10Comparison of India with China on several relevant features related to traffic safety. (Data are

for 2006 and 2007.)

Aspect India China

Gross nationalincome per capita,US$

820 2,000

Gross nationalincome per capita,US$ purchasing

power parity

2,460 4,660

Vehicle populationCars: 13%

Motorized two-wheelers: 72%

Cars: 14%

Motorized two-wheelers: 55%

Traffic fatalities

Annually 95 per million persons

National distribution by road usernot known; estimates indicate

pedestrians, bicyclists andmotorized two-wheelers: > 60%

Children (0-14): 7%

Elderly (> 64): 8%

Male to female ratio: 5.4

Annually 68 per million persons

Pedestrians, bicyclists, andmotorized two-wheelers: 59 %

Children (0-15): 6%

Elderly (> 65): 1%

Male to female ratio: 3.2

Regulations

Seat belt use mandatory, but manystates not enforcing the law. Whereenforced, seatbelt use is ~70%.

Helmet use mandatory, but manystates not enforcing the law. Whereenforced, helmet use > 90% in thedaytime.

Daytime headlight use is notmandatory for any type ofmotorized vehicle.

Law against drinking and drivingon the books, but not enforcedstrictly.

Seat belt use mandatory, but many provinces are not enforcing the laweffectively. Where enforced, seat

belt use ~70% in the daytime.

Helmet use is not mandatory.

Daytime headlight use is notmandatory for any type ofmotorized vehicle, except in fog.

Zero alcohol tolerance, strictlyenforced in urban areas.

Vehiclecrashworthiness

By 2012, all cars sold in India willhave to satisfy crashworthinessregulations similar to those inEurope.

No specific regulations, but de facto required for marketing due tocustomer demand.

RoadsExpressways: < 500 km.

No separate lanes for bicyclists inurban areas.

Expressways: > 40,000 km.

Dedicated bicycle lanes in manyurban areas.


56/62

51

Table 11Highlighted major problems areas in India (this report) and in China (Zhang et al., 2008).

India China

Pedestrians and other non-motorist in urbanareas

Pedestrians, other non-motorists, and slowvehicles on highways

Pedestrians and other non-motorists

Motorcycles and small cars in urban areas Motorcycles

Over-involvement of trucks

Nighttime driving Nighttime driving

Wrong-way drivers on divided highways

Vehicle passengers


57/62

52

SUMMARY

The present report was designed to analyze the traffic safety situation in India, and to

identify countermeasures that would address areas in which the total harm caused by crashes can

be substantially and readily reduced. The report focused on two aspects of traffic safety in India,

challenges and opportunities. The first part of the report provided a comprehensive analysis of

the current traffic safety situation in India. It was pointed out in this analysis that fatality rates

have increased both on highways and in urban areas during the past few years. Theoretical

models suggest that the number of fatalities in India is not likely to start to decline for many

years to come unless new policies are implemented. Based on the present analysis, the followingsix areas were identified as having potential for substantially reducing fatalities in India: (1)

pedestrians and other non-motorist in urban areas, (2) pedestrians, other non-motorists, and slow

vehicles on highways, (3) motorcycles and small cars in urban areas, (4) over-involvement of

trucks and buses, (5) nighttime driving, and (6) wrong-way drivers on divided highways. The

second part of the report outlined several promising countermeasures for each of these six areas.

The third part of the report presented a brief comparison of major traffic safety challenges in

India and China.

There is an urgent need to revamp police data collecting procedures so that necessary

information is available for scientific analysis. India specific countermeasures will be possible

through continuous monitoring and research, which will require the establishment of road safety

research centers in academic institutions. The proposed National Road Safety Board would be

the obvious agency that could help move toward a safer future as outlined above.


58/62

53

REFERENCES

Cather, C. (2007). Faster, better, cheaper & cleaner: The dynamics driving the global

automotive industry . Delhi: CSM Worldwide.

Census of India. (2008). Projected population by sex as on 1st March . Retrieved September 20,

2008, from http://www.censusindia.gov.in/Census_Data_2001/Projected_Population/

Projected_population.aspx

CRTAS. (2007). China road traffic accident statistics. Beijing, China: Traffic Administration

Bureau of China State Security Ministry.

Committee on Infrastructure. (2007). Report of the Committee on Road Safety and Traffic

Management . Retrieved July 22, 2008, from

http://infrastructure.gov.in/pdf/sunder_report.pdf

Committee on Infrastructure. (2008). Highways . Retrieved July 21, 2008, from

http://infrastructure.gov.in/highways.htm#b1

Consulting Engineering Services. (2007). Road safety assessment report . Chandigarh, India:

Punjab Roads & Bridges Development Board.

CRRI [Central Road Research Institute] (2007). Annual report 2007 . New Delhi: Author.

Department of Road Transport and Highways (2008a). Motor transport statistics . Retrieved

July 21, 2008, from http://morth.nic.in/index2.asp?sublinkid=202&langid=2

Department of Road Transport and Highways (2008b). Annual report 2007-2008 . New Delhi:

Ministry of Shipping, Road Transport and Highways.

Department of Road Transport and Highways (2008c). Basic road statistics . Retrieved July 21,

2008, from http://morth.nic.in/index2.asp?sublinkid=157&langid=2

Department of Road Transport and Highways (2008d). Traffic data . Retrieved August 24, 2008,

from http://morth.nic.in/index2.asp?sublinkid=369&langid=2


59/62


60/62

55

McManus, W. (2007). The economics of road safety: An international perspective (Technical

Report No. UMTRI-2007-23). Ann Arbor: The University of Michigan Transportation

Research Institute.

Ministry of Shipping, Road Transport and Highways. (2006). State-wise number of persons

killed in road accidents on national highways, state highways and other roads of India

(2004) . Starred Question No. 169. New Delhi: Parliament of India, Rajya Sabha.

Ministry of Statistics and Programme Implementation. (2008). Statement S-1.1: Macro

economic aggregates and population (at current prices) . Retrieved June 26, 2008, from

http://mospi.nic.in/mospi_nad_main.htm

Mishra, B. K., Banerji, A. K., and Mohan, D. (1984). Two-wheeler injuries in Delhi, India: A

study of crash victims hospitalized in a neuro-surgery ward. Accident Analysis and

Prevention, 16 , 407-416.

Mohan, D. (2004). The road ahead: Traffic injuries and fatalities in India. New Delhi: Indian

Institute of Technology Delhi, Transportation Research and Injury Prevention

Programme.

Mohan, D. and Tiwari, G. (2000). Road safety in less motorised countries - Relevance of

international vehicle and highway safety standards . International Conference on Vehicle

Safety 2000, pp. 155-166. London: Institution of Mechanical Engineers.

National Sample Survey Organisation. (2005). Employment and unemployment situation in

India (JanuaryJune 2004) . New Delhi: Ministry of Statistics and Programme

Implementation.

NCRB [National Crime Records Bureau]. (2007). Accidental deaths & suicides in India - 2006 .

New Delhi: Ministry of Home Affairs, National Crime Records Bureau.

NHTSA [National Highway Traffic Safety Administration]. (2008). Traffic safety facts 2006

data . Retrieved August 5, 2008, from http://www-nrd.nhtsa.dot.gov/Pubs/810809.PDF


61/62

56

Planning Commission (2006). Towards faster and more inclusive growth: An approach to the

11th five year plan (2007-2012) . New Delhi: Government of India.

Registrar General & Census Commissioner India. (2008). Census of India 2001 . New Delhi:

Office of the Registrar General & Census Commissioner India. Retrieved August 5,

2008, from http://censusindia.gov.in

Saija, K. K. and Patel C. D. (2002). Micro level study of accidents on NH-8 passing through

Valsad District. Indian Highways, 30 , 43-51.

Saija, K. K., Patel, C. D., and Upadhyay, V. J. (2000). Spectrum analysis of road accidents.

Indian Highways, 28 , 29-41.

Shaheem, S. and Das Gupta, G. C. (2005). Impact of road development on road safety - A case

study of Aluva-Cherthala section of NH-47. Journal of the Indian Roads Congress, 66 ,

615-639.

Shaheem, S., Mohammed, K. M. S., and Rajeevan (2006). Evaluation of cost effectiveness of

improvements of accident prone locations on NH-47 in Kerala state. Indian Highways,

34, 35-46.

Shrinivas, P. L. L. (2004). Studies undertaken to identify critical causes of accidents in the

highways of Tamil Nadu. Indian Highways, 31 , 11-22.

Sivak, M. and Tsimhoni, O. (2008). Improving traffic safety: Conceptual considerations for

successful action. Journal of Safety Research , 39, 43-457.

Smeed, R. J. (1972). The usefulness of formulae in traffic engineering and road safety. Accident

Analysis and Prevention, 4 , 303-312.

Society of Indian Automobile Manufacturers. (2008). New Delhi. Personal communication.

Stevenson, M., Yu, J., Ying, Z., Hendrie, D., Ivers, R., Li, L-P., and Norton, R. (2007). China

seat belt intervention . Sydney, Australia: The George Institute for International Health.


62/62

Thulin, H. and Nilsson, G. (1994). Road traffic, exposure, injury risk and injury consequences

for different travel modes and age groups (Technical Report No. 390A). Linkping:

Swedish Road and Transport Research Institute.

Tiwari, G., Mohan D., and Gupta, D. P. (2000). Evaluation of capacity augmentation projects of

national highways and state highways . New Delhi: Ministry of Surface Transport.

Tiwari, G., Mohan, D., and Fazio, J. (1998). Conflict analysis for prediction of fatal crash

locations in mixed traffic streams . Accident Analysis and Prevention, 30 , 207-215.

Tiwari, G., Mohan, D., and Muskaug, R. (1998). Mumbai urban transport project 2: Accident

study . New Delhi: Indian Institute of Technology Delhi.

UNICEF. (2007). The state of the world's children 2008: Child survival . New York: Author.

U.S. Census Bureau. (2008). Table 2: Annual estimates of the population by sex and selected

age groups for the United States: April 1, 2000 to July 1, 2007 . Retrieved August 5,

2008, from http://www.census.gov/popest/national/asrh/NC-EST2007-sa.html

Varghese, M., and Mohan, D. (2003). Transportation injuries in rural Haryana, North India. In

Proceedings of the International Conference on Traffic Safety , pp. 326-329, New Delhi.

World Bank. (2008). 2008: World development indicators . Washington D.C.: International Bank

for Reconstruction and Development.

Zhang, W., Tsimhoni, O., Sivak, M., and Flannagan, M. J. (2008). Road safety in China:

Challenges and opportunities (Technical Report No. UMTRI-2008-1). Ann Arbor: The

University of Michigan Transportation Research Institute.

Documents

ROAD SAFETY IN INDIA: CHALLENGES AND OPPORTUNITIES